Solution-based synthesis of CsSnI3

Synthesis of CsI Solution

12 gram of CsI powder (99.9% purity) was added to 30 gram of deionized water to form 28.6 wt % (weight percent) CsI solution. The CsI powder was fully dissolved in deionized water. The CsI solution was stirred for 30 minutes.

CsI solution was colorless and stable in ambient conditions.

The CsI solution could comprise more CsI powder as long as the amount of CsI powder was fully dissolved in deionized water, and the CsI solution was colorless. CsI powder could dissolve in less than its own mass of water without apparent decomposition.

It would be apparent to one skilled in the art that CsI solutions could be made using any solvents other than those used in the examples. Examples of the solvent include but not limited to, water, distilled water, mixtures thereof.

The wt % range of the CsI solution was ranging from about 5.0 wt % to about 50.0 wt %, or from about 15 wt % to about 40 wt %.

Synthesis of SnCl 2 Solution

An amount of 4 gram of SnCl 2 powder (99.9% purity) was added to 20 gram of anhydrous ethanol to form 16.7 wt % of solution. The SnCl 2 powder was fully dissolved in anhydrous ethanol to form the SnCl 2 solution. The solution was stirred for 30 minutes.

SnCl 2 solution was colorless and stable in ambient conditions.

The SnCl 2 solution could comprise more SnCl 2 powder as long as the amount of SnCl 2 powder was fully dissolved in anhydrous ethanol, and the solution was colorless.

It would be apparent to one skilled in the art that SnCl 2 solutions could be made using any solvents other than those used in the examples. Examples of the solvent include but not limited to, anhydrous alcohol, such as but not limited to, anhydrous methanol and anhydrous isopropyl; glacial acetic acid; ethylene glycol or triethylene glycol; and the mixtures thereof.

The wt % range of the SnCl 2 solution was ranging from about 20 wt % to about 80%, or from about 10 wt % to about 50 wt %.

Synthesis of CsSnI 3 Film by Drop-Coating

A given amount of the prepared CsI solution was transferred to a reaction vial first. SnCl 2 solution was then slowly dropped into the vial by a pipet. The ratio of CsI solution to SnCl 2 solution was in a range of about 1/10 to about 1/3. CsSnI 3 was formed immediately as black grains suspending in the mixed solution. This phenomenon implied that the CsSnI 3 was neither soluble in DI water nor in ethanol.

After five minutes sonication (using ultrasonic vibrator), homogeneous black CsSnI 3 solution with CsSnI 3 black grains uniformly distributed in the solution, was formed. The homogeneous CsSnI 3 solution contained CsSnI 3 grains with a typical diameter of a few micrometers.

The homogeneous CsSnI 3 solution was drop-coated on a substrate, such as a microscope glass slide or a piece of ceramics.

After drop-coating, the substrate was heat treated at 70° C. for 5 minutes to remove residual solvents. SnCl 2 film with a thickness ˜40 μm was obtained by drop-coating the homogeneous CsSnI 3 solution. The heating temperature ranged from about 50° C. to about 90° C.

The cross-sectional scanning electron micrographic (SEM) image taken from the film on a glass substrate was shown in FIG. 2 . FIG. 2 indicated that the film was dense and filled with large polycrystalline domains ranging from a few micrometers to 10 μm.

The chemical reaction for the mixture of the solutions could be described as the following:

3CsI+SnCl 2 →CsSnI 3 +2CsCl,

The reaction was verified by identifying the end products of CsSnI 3 and CsCl using the X-ray diffraction (XRD) data to be discussed later.

Photoluminescence (PL) data and X-ray diffraction data (XRD) were used to identify CsSnI 3 and to characterize the crystalline quality of each individual domain.

PL spectra were taken from a Nanolog system from Horiba Jobin Yvon. The Nanolog system had a 450 W Xe-lamp light source, a double-grating excitation spectrometer to select a central excitation wavelength and its bandwidth, a sample compartment, and an emission spectrometer to spectrally select desired emission to a photomultiplier tube (Hamamatsu P2658P) coupled with single photon-counting electronic circuits.

A relative low photoexcitation level of ˜20 mW cm −2 was used.

As shown in FIG. 3 , the PL intensity of a typical solution-based drop-coated CsSnI 3 film on a glass substrate was about half of the PL intensity taken from a typical thin-film synthesized from thermal evaporator powders by the vacuum-powder based method (as disclosed in U.S. patent application Ser. No. 13,151,243) under the identical excitation and PL collection conditions.

The PL line shape was identical for the solution-based and vacuum-powder based deposition or synthesis methods as displayed in the inset of the FIG. 3 . This result strongly suggested that the CsSnI 3 film quality was compatible using two different synthesis methods.

XRD data were used not only to further verify the crystal structure of CsSnI 3 but importantly also to identify the side products when the CsI solution was formed as the end products and mixed with the SnCl 2 solution (as shown in the reaction).

FIG. 4( a ) showed the XRD data measured from the same drop-coated CsSnI 3 film used for taking the SEM image in FIG. 2 .

FIG. 4( b ) showed theoretical calculated XRD profile by using a commercial software package CASTEP written by researchers at the Universities of York, Durham, St. Andrews, Cambridge and Rutherford Labs, using orthorhombic crystal structure of CsSnI 3 . CASTEP uses density functional theory with a plane wave basis set to calculate the electronic properties of crystalline solids and other materials from first principles.

All the measured peaks were accounted for and sorted into four groups.

The first group consisted of the peaks of 1 to 3 and all others not number-labeled. These peaks were originated from the orthorhombic structure of CsSnI 3 and match the calculated XRD profile, as taught by CASTEP, as shown in FIG. 4( b ).

Peaks 1 and 2 reflected the expected XRD features of the Sn—I—Sn bond tilting) (163° in the a- and b-direction, respectively, in reference to the cubic-structure of CsSnI 3 .

Peak 3 represented the signature of the Sn—I—Sn bond tilting (170°) in the c-direction in reference to the tetragonal structure of CsSnI 3 .

The second group included the peaks of 4 and 5. Those peaks were originated from the residual material CsI.

The peaks of 6 and 7 made up the third group. The peaks of 6 and 7 were originated from CsCl, leading to the unambiguous determination of the chemical reaction as the CsI solution was transferred into the SnCl 2 solution to form CsCl.

Finally, the presence of peaks 8 and 9 made up the fourth group. The peaks 8 and 9 matched Sn (IV) dioxide, SnO 2 . It should be mentioned here that the absence of XRD peaks associated with the residual SnCl 2 strongly suggested that the chemical reaction under the ambient conditions resulting in favor of CsSnI 3 formation.

In summary, CsSnI 3 were synthesized using the CsI and SnCl 2 solutions.

A solution-based method, capable of large-scale production, was employed to fabricate CsSnI 3 films at a low cost, especially suitable for solar cell applications. The film polycrystalline quality was characterized by SEM, PL, and XRD data. Those data was verified to be comparable to that of similar thin-films synthesized by the vacuum-powder-based evaporation method.

While the invention has been described in detail and with reference to specific examples and the embodiments thereof, it will be apparent to one skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope thereof.